Artificial intelligence may be aviation’s most promising partner
AI and related technologies yield greater safety, improved efficiency, and enhanced performance, while streamlining processes and reducing the risk of human error.
By Don Van Dyke
ATP/Helo/CFII. F28, Bell 222
Pro Pilot Canada Technical Editor
The Federal Aviation Administration (FAA) and the European Union Aviation Safety Agency (EASA) are responsible for overseeing the safety of commercial aviation, including the integration of AI technologies in avionics systems, ensuring that AI systems used in aviation are secure and do not pose cybersecurity risks. EASA released its AI Roadmap 2.0 in 2023 to further integrate AI into aviation with a human-centric approach. Recently, the Biden-Harris administration announced new AI actions, receiving major voluntary commitment on AI innovation.
Artificial intelligence (AI) is the capability of computer systems or algorithms to imitate intelligent human behavior. AI encompasses a wide range of technologies and applications, including robotics, natural language processing, and computer vision.
AI-powered systems can aid aviation by automating repetitive operational assignments, such as flight planning, crew scheduling, and inflight services. They can also prove helpful in functions related to maintenance, repair, and overhaul (MRO), like aircraft maintenance scheduling, compliance checks, inventory control, and training – all while minimizing errors and increasing efficiency. These innovations serve to make all sectors of air transport safer, more efficient, and increasingly client-oriented.
Understanding AI
AI is an evolution from rule-based technology (eg, IF “condition” THEN “result”) in pursuit of smarter, more intuitive technologies which increase the capabilities or reduce the workload of humans in complex scenarios.
Machine learning (ML) is a subset of AI that enables a computer to learn to perform tasks by analyzing a large dataset and generalizing to unseen data, thus performing tasks without explicit programming. It deals with the creation of models that can improve their performance over time as they are exposed to more data.
ML is the primary way that most people interact with AI. These interactions happen most commonly when receiving video recommendations on an online video streaming platform; troubleshooting a problem online with a chatbot which directs you to appropriate resources based on your responses; or interacting with virtual assistants who respond to your requests to schedule meetings in your calendar, play a specific song, or call someone.
While all ML is a form of AI, not all AI involves ML. Simply, AI is the overarching concept of machines mimicking human intelligence, and ML is one method used to achieve that goal.
Rapidly developing generative AI (GenAI) can create new content (eg, imagery, text, audio, synthetic data) in response to a submitted prompt, such as a query, by simulating the learning and decision-making processes of the human brain to produce new and relevant content.
Simply, AI analyzes data and patterns from which it makes predictions and formulates recommendations. GenAI learns from AI output and generates or substitutes other data to innovate patterns and perspectives, thereby offering new insights, predictions, and recommendations.
When organizations begin experimenting with GenAI, they often underestimate its transformative nature and the operational challenges it entails. Despite great promise, a recently-released Monmouth University poll found that only 9% of Americans believe that computers running AI would benefit society rather than harm it.
Digitalization of aviation seeks to steer innovation responsibly, aligning progress with enduring human values. For high-consequence, low-probability endeavors, like aviation, exacting diligence in governance, control, and implementation is crucial to improving public confidence in the value of AI.
Chatbots
A chatbot is a program that engages in human conversation using AI and natural language processing (NLP) to understand and respond to questions or requests. Beyond simple questions and answers, chatbots are evolving into conversational companions, enhancing customer service, and streamlining interactions. As they become more sophisticated, the line between human and AI communication blurs, reshaping the way we interact.
The role of data
Responsible collection, management, and utilization of data is fundamental to unleashing the full potential of AI. Data fuels algorithms, shapes insights, and supports predictions. AI can analyze vast amounts of data to provide insights and support decision-making processes, from route planning to market analysis. AI can simulate various operational scenarios to help management make informed decisions.
Table 1 presents an overview of common aviation software applications of AI/ML/GenAI. Table 2 identifies areas of concern in implementing AI/GenAI, and highlights needs to both understand and control aspects of this information technology.
Implementing AI/GenAI
The minimum size for an aviation company to consider implementing AI varies, but generally, even small to mid-sized companies can benefit from AI technologies when pursuing organizational goals, such as operational efficiency, cost savings, customer experience, and/or competitive advantage. Companies can start with AI applications in areas like predictive maintenance and client service to realize immediate benefits.
The following key steps are recommended when implementing an AI system:
• Set clear objectives. Whether improving operational efficiency, enhancing customer experience, or driving innovation, aligning AI initiatives with strategic goals is essential for success.
• Ensure data quality and privacy. Establish robust governance to ensure data quality, privacy, and integrity to build trust with customers and stakeholders.
• Promote a learning culture. Encourage continuous learning and upskilling within the organization, fostering collaboration between data, operational, and business units.
• Consider ethics. Address ethical concerns associated with AI by providing relevant frameworks to guide its development and use.
• Start small, scale cautiously. Begin with pilot projects to validate AI concepts and demonstrate tangible value. Scale up gradually, leveraging lessons learned and iterating continuously to improve outcomes.
Future trends
The future of AI/GenAI will herald quantum advances in ethical considerations and novel applications. From service innovations to societal transformations, understanding emerging trends will be increasingly essential to staying at the forefront of this technology. For instance, AI/GenAI can improve predictive maintenance by analyzing sensor data and failure patterns to predict when maintenance is required.
It can also create virtual models of aircraft components to simulate wear and identify potential issues before they occur. By generating detailed repair and troubleshooting guides, it can assist technicians in performing complex repairs more efficiently. And GenAI can generate and update technical manuals and training materials, ensuring that maintenance staff have the most current information.
Furthermore, supply chain can be optimized using GenAI to analyze data from suppliers, logistics providers, and customers in order to reduce costs and improve efficiency. Finally, it can improve client service by providing fast and accurate responses to client inquiries.
While AI/GenAI augment capabilities, the human touch – empathy, creativity, and intuition – remains irreplaceable. Nurturing these synergies ensures that AI/GenAI becomes a powerful ally, enhancing our endeavors without diminishing the unique qualities that make us human.
The most pragmatic approach is to embrace the opportunities offered by AI/GenAI, navigate its ethical complexities, and envision a future where, aligned with human values, it propels us toward new frontiers of innovation.
Don Van Dyke is professor of advanced aerospace topics at Chicoutimi College of Aviation – CQFA Montréal. He is an 18,000-hour TT pilot and instructor with extensive airline, business and charter experience on both airplanes and helicopters. A former IATA ops director, he has served on several ICAO panels. He is a Fellow of the Royal Aeronautical Society and is a flight operations expert on technical projects under UN administration.